Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Missing data in model's labels.csv for DICOM-SEG conversion #80

Open
ivantyj opened this issue Sep 27, 2024 · 2 comments
Open

Missing data in model's labels.csv for DICOM-SEG conversion #80

ivantyj opened this issue Sep 27, 2024 · 2 comments

Comments

@ivantyj
Copy link

ivantyj commented Sep 27, 2024

Thank you for your great work!
I’m working on converting the output to DICOM-SEG format by creating metadata.json for itkimage2segimage using labels.csv.

I noticed some missing data in labels.csv:

  • For the abdominal-organs-v2.0.0 model, the entry 17,thyroid_gland,SCT ... is missing.
  • For the whole-body-v2.0.1 model, on line 63, the value for SegmentedPropertyTypeModifierCodeSequence.CodingSchemeDesignator should be SCT.

I’ve only checked these models and body-v2.0.0, so other models haven’t been reviewed yet.

@lassoan
Copy link
Owner

lassoan commented Sep 27, 2024

Thanks a lot for the review, these are very useful. Would you mind sending the files with the proposed changes (upload somewhere and post the link)?

Instead of converting to metadata.json, you might consider updating itkimage2segimage ro accept labels.csv directly. Json is great for developers, but I find many people are unable to create metadata files in json, while they are comfortable with putting together a csv file in Excel. We plan to further simplify the labels.csv format by shortening the column names (in the next couple of weeks).

@ivantyj
Copy link
Author

ivantyj commented Oct 16, 2024

Thanks a lot for the review, these are very useful. Would you mind sending the files with the proposed changes (upload somewhere and post the link)?

Sorry for the late reply. The updated .csv file is here:
whole-body-v2.0.1_labels.csv
abdominal-organs-v2.0.0_labels.csv

Instead of converting to metadata.json, you might consider updating itkimage2segimage ro accept labels.csv directly. Json is great for developers, but I find many people are unable to create metadata files in json, while they are comfortable with putting together a csv file in Excel. We plan to further simplify the labels.csv format by shortening the column names (in the next couple of weeks).

It's a good idea to have itkimage2segimage accept labels.csv. However, I'm not sure if the information in labels.csv is sufficient to create a DICOM-SEG file. I've hardcoded several properties (e.g., ContentCreatorName, ContentDescription, ClinicalTrialSeriesID, etc.) in metadata.json, and I'm unsure what effect that might have.

I've only tested some converted DICOM-SEG files in 3D Slicer. The segments load correctly, but the recommendedDisplayRGBValue is not displaying as expected.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants